The Origami Forum

wolf wrote:...and therein lies the drawback of relying on the origami database as a sole source - it's not yet a comprehensive listing of what's out there!

...and the strength of the database is that it can be added to by anyone at all. Anyone who, for example, knows of kangaroo diagrams that aren't listed in the database could simply add the model using the 'Add model to Database' link.

www.origamidatabase.com/addmodel.asp

...and the strength of the database is that it can be added to by anyone at all. Anyone who, for example, knows of kangaroo diagrams that aren't listed in the database could simply add the model using the 'Add model to Database' link.

yes, and that's also the biggest weakness. if the diagrams of webpages are includes, there's always a very big risk, that the webpage may be offline, or have a changed URL. the diagrams could also be deleted and so on. it's better, not to include the web-diagrams into the database!


Christian

TheRealChris wrote: yes, and that's also the biggest weakness....

It's the biggest weakness of the web, not the database. Anything that catalogues web based items or contains links has this problem.
A simple email in the comments/corrections will cover URL changes etc.

Also, my comment was intended to cover books, magazines etc., not just websites.

however, we're a bit off-topic here

It's the biggest weakness of the web, not the database. Anything that catalogues web based items or contains links has this problem.

right, but we would take over this problem into a good working system

A simple email in the comments/corrections will cover URL changes etc.

just let three or four years pass, and the number of deadlinks will rise to the sky. who will check the links? and who will edit them? maybe too much effort? what's the use of an diagram-link, that may already be dead? I mean, there would probably be hundreds of links to hundreds of maybe-online-diagrams. who will keep an eye onto those maybe-online-diagrams. and who will find the new webpage, when the database link isn't working?

however, we're a bit off-topic here

so I'll stop here

Chris, why don't you split off the last few posts and put it into the Model Database section? It might generate some useful discussion on how to improve the database, particularly for diagrams hosted on websites.

good idea... done


Christian

Thanks for the suggestions guys, I'll pass these to Dennis (aka Database man). What I've always recommended for website additions is only linking to the root folder/site, as opposed to 'hot linking' directly to the source PDF/Image file. This way, if the website does ever get redesigned, a user should be able to find the diagram from the Home page of the website.

Saj

Thanks, Chris.

Back to the problem of online diagrams:

I think online diagrams should be catalogued into a database of some sort, but perhaps it should be kept separate from the book/magazine database, because of intransient nature of web links.

I realise that you can already limit the search to just websites, but another reason for doing so is the link addition process. Right now, the "Add Model" routine is mostly tailored to books and magazines; it's also a fairly involved form to fill in (the drop down menu in the "book/model checking" routine takes forever to load!). After which it has to go past the editors before being added to the database. It's a robust system involving many checks, which is good because it maintains the high quality of the database.

For websites however, is it really worth the submitter's/editor's effort to process something which could go obsolete the next day, month or year? Plus, one would have to routinely do link checking and updating to ensure that the database is kept up to date.

A Wiki-style, user-editable database for online diagrams might be the way to go. There's always the risk of Wiki abuse, but this can be kept to a minimum by moderators, log-in only editing, edit history, etc. The question is whether there's a large enough userbase to make this successful.

While we're at it, maybe we should also set up an OriWiki...

A wiki (MediaWiki for example) for the online models is a great idea !

It is very easy to create categories , sections, subsections etc, and to add/edit the pages...

wolf wrote: Right now, the "Add Model" routine is mostly tailored to books and magazines; it's also a fairly involved form to fill in (the drop down menu in the "book/model checking" routine takes forever to load!). After which it has to go past the editors before being added to the database. It's a robust system involving many checks, which is good because it maintains the high quality of the database.

I agree that it is a little involved. The long load is simply to try to avoid duplicate entries, buut we could probably simplify this.

wolf wrote: For websites however, is it really worth the submitter's/editor's effort to process something which could go obsolete the next day, month or year? Plus, one would have to routinely do link checking and updating to ensure that the database is kept up to date.

My arguments here have always been that firstly, websites don't actually change that often, and even if they did, then only the base name will change, not the contents, in which case a simple email will fix it. Secondly, I've also tried to get people to keep their own website entries up to date, but with little success.

wolf wrote: A Wiki-style, user-editable database for online diagrams might be the way to go. There's always the risk of Wiki abuse, but this can be kept to a minimum by moderators, log-in only editing, edit history, etc.

This is pretty much what the Origami Database is! A moderated, user-editable database. The problems arise with making such a thing searchable in a sensible way. At this point you need specific fields and names and formats or it just ends up being a mess.

I'll have another look at making it easier/faster to add data (like skipping the list!) and maybe add specific 'Add Website' and 'Search Website' functions. But give me until the New Year to have it ready

denori wrote:I agree that it is a little involved. The long load is simply to try to avoid duplicate entries, but we could probably simplify this.

Would it be possible to add the duplicate checking after the submission process? Basically, all submissions would be placed in a queue that can be processed at a later stage. I can see that having the duplicate checking during submission will prevent submitters from unknowingly adding duplicate entries (and hence wasting their time), but perhaps the onus of checking for duplicates should be placed on the submitter. Perhaps a big "CHECK THIS LIST BEFORE SUBMITTING!" sign at the submission page? Submitters can then ignore it at their own peril...

denori wrote:My arguments here have always been that firstly, websites don't actually change that often, and even if they did, then only the base name will change, not the contents, in which case a simple email will fix it.

It's more the dead link problem, and knowing netizens, sending simple emails is easier said than done. If you could automatically feed the WWW entries into a linkchecking site, that could help prune down the number of dead entries.

denori wrote: Secondly, I've also tried to get people to keep their own website entries up to date, but with little success.

Guilty as charged. My excuse is that sometimes it's not the website owner that does the submission for their sites, but a third party. So website owners may not even know that their entries have been placed on the database.

denori wrote:This is pretty much what the Origami Database is! A moderated, user-editable database. The problems arise with making such a thing searchable in a sensible way. At this point you need specific fields and names and formats or it just ends up being a mess.

I was thinking something bigger, really, more like a comprehensive online encyclopedia of origami history, tips, tricks, techniques, what-have-you. The origami model database would then be a subset of this, and the wiki search options can be configured to search the database only, if the user so wishes. As it is now, I think the search on the ODB works nicely; often you don't get too many returned results that it's impractical to look them over by eye.

Having said all that, I'll now say that IMO, the ODB has been the most valuable addition to the online origami world over the last few years. I still see it as having teething problems, but even so, nothing else out there provides a comparable service, so you guys at ODB definitely deserve a big thank you for all your efforts!

I guess a Wiki of some sort would be cool - much like the online WikiPedia.

The database is cool - many thanks to Dennis for the ongoing maintanance (and of course the rest of the ODB team). When I have some time over Xmas I might moderate some more of the entries we receive.

Saj
PS - Dennis - any chance you'll be coming to the convention in Leicester?

Thanks for the kind words
If a Wiki were to be started, then one good place to start would probably be (I think ) Dorothy Englemann's site. I think it was she who started the 'Encyclopaedia Origamica'. Then there's the wealth of knowledge contained in David Lister's stuff on the BOS site.
There was an attempt to include biographical info in the ODB (It can still be done!) but this would be better farmed out to a Wiki, as would the putative origami calendar I started. That way the Model database would be remodelled and the focus concentrated on books, websites etc. and the models contained.

If you want to see what I was hoping for, try clicking on the name of a folder to get their info. The only one with any real data on it is mine, so click on my name!

Saj, I haven't booked yet, but I'm coming to Leicester! See you there!

Wolf,

If you could automatically feed the WWW entries into a linkchecking site, that could help prune down the number of dead entries.

I didn't know that could be done! Can you email me privately with some details please? Thanks.

As for websites being added 'by a third party'. It was usually me
Every now and then I'd find something and just add it, just to keep the database ticking over.

denori wrote:If a Wiki were to be started, then one good place to start would probably be (I think ) Dorothy Englemann's site. I think it was she who started the 'Encyclopaedia Origamica'. Then there's the wealth of knowledge contained in David Lister's stuff on the BOS site.

All of these could be incorporated into a Wiki and then built upon by subsequent Wiki users. I think it might be an interesting experiment to try, does anyone out there want to volunteer to set it up?

denori wrote:I didn't know that could be done! Can you email me privately with some details please? Thanks.

I've sent you a private message here with more info on this.